The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the gemini CLI tool, which is described as having the capability to run shell commands and edit files. The instructions recommend using the --yolo flag to auto-approve all tool calls, which bypasses manual review of potentially dangerous operations.\n- [PROMPT_INJECTION]: The skill is designed to ingest and process large amounts of untrusted data from external sources, creating a surface for indirect prompt injection.\n
Ingestion points: Processes large codebases, external APIs via Google Search, and web content fetched at runtime, as described in SKILL.md.\n
Boundary markers: No boundary markers or instructions to ignore embedded commands are present in the delegation prompts or CLI usage patterns.\n
Capability inventory: The gemini tool can read/edit files, run shell commands, and fetch web content, providing a high-privilege execution environment for potentially injected instructions.\n
Sanitization: There is no evidence of sanitization for the data processed by the external CLI or the output captured and used for subsequent tasks.

gemini-cli